@mkj256 It was in M17 so that sunspot link won't help you then.

You could follow the links here for the only other two power spectrum links.  They may be of help

http://www.mapleprimes.com/questions/146021-Spectrum-Analysis-Of-A-PWM-Signal-To#answer146027

I made a contribution for Maple in 2011 under Constrained random points on a circle. 

I wonder how this answer fairs today with Maple 18 ?

Of course you would want the supported hardware Nvidia and not the AMD Radeon (it's up to AMD to implement CUDA compatability for their cards).  So you would at least want GPU hardware supported by CUDA. 

I'm not an expert in this field but in general Nvidia graphics cards are the ones you should stick to.

Default works but the points are too large so I had thought if I set symbol=point they would automatically be set to symbolsize=1.  I should mention that I had forgot to say when I set symbolsize=1 I left symbol unspecified as you mention above, thanks for clarifying that Acer.

It must have been fixed in 18.01.  It doesn't need to be so many points.  The three points from the help page is all that's needed to see if it works.  So it must be an 18.00 issue only.

Converting the pointplot3d help page to a worksheet for the first example changing symbol to point .. It works in M12 but fails to display in M18.

@descartes0000

When you say zoom, do you mean dynamic zooming (ie see finer and finer details as you zoom in)?

@Mac Dude thanks I'll give that a try. 

@Jakubi - thanks for the -X options, probably I set the -X options in the .ini file - not too sure

Graph window sizing seems to have a huge impact on performance.  I resized the Spectrogram graph back to the larger size and Mem usage shot up to ~1,500,000 took a little while before I had control again.  Hiding the window by closing that subsection then minimizing and maximizing the Maple window gave me the  memory back.  Everything is slower now.  Physical memory described to me in the windows task manager show 2.69Gb of memory used and the graph window in Maple still hasn't resized.  I have killed the process to Maple.exe

@Mac Dude The limitjvmheap is false yes.  I will have to try in the morning as I am not near Maple.  I will couble the stacklimit and heap size and see what happens. 

Still need to double check the memusage in the same session ie/ Open new worksheets in seperate tabs, run up the memusage in the task manager close tabs and see if the memory is released.  Maybe if I am able to restart on the same tab, the memory is released.  Although I am wondering that maybe when one worksheet runs up the memory and is closed without restart on the same sheet the memory is not released ... I'll have to check later.

@Carl Love I'm referring to the maxheap from the launch.ini file in the bin.win folder of the Maple directory.  I am not sure if the value set in the launch.ini file is the same as a value set in kernelopts.

my jvmheaplimit is currently at 65536

@Christopher2222 actually setting maxheap=2000m will not start. 

Actaully I didn't think the maxheap size was a hardcoded limit if it was not set.  I thought it was a floating value.  Anyways I will see if the 1600m limit helps.

I'm sure sending the image to a file will definitely improve the response as Mac Dude has suggested. 

Thanks for all the insights.  I will be able to check later and do some more experimenting.

I do recall a while ago that in order to avoid some java issues one would have to add the line maxheap=700m in a maple.ini file in the main Maple directory but then I recall at some later time that it wasn't required anymore.

You mention increasing the limit to 2Gb, so will adding maxheap=2000m to maple.ini do the trick?

**edit ** I meant the launch.ini file

@Carl Love Yes I am referring to the task manager values.  I am not near Maple right now but I didn't actually record what Maple was reporting in it's window pane, only what windows was reporting in the task manager.  After restart there was no release of memory in the Windows task manager - it remained high.  Perhaps it takes a little more time to release the memory, next time I have a chance I will wait 10 minutes before closing the mserver.exe and maple.exe processes.

@Alejandro Jakubi I agree no stackexchange similarities here please.

On the old primes there were some interesting blogs which now are converted to posts, and unfortunately the child pages are no longer linked with them.  Tools for the Maple Masters and High performance programming techniques have links but are not linked with any other pages (old primes had them linked to other pages). 

Also poll questions gave another dimension of interactivity with this site.  Perhaps administrators could bring it back?

@Alejandro Jakubi Not meant to be a mindset definintion.  More so as a vague comparison.